Once he tasted some chamomile tea and sent it back, saying only: “The stuff tastes of window.” Both she and the servants were surprised because they had never heard of anyone who had drunk oiled window, but when they tried the tea in an effort to understand, they understood: it did taste of window.
– Gabriel Garcia Marquez

Though perhaps not the most scientific or accurate account of synesthesia, this quote provides a slight intuition to the nonsynesthete of what it might be like to have the condition. Synesthetes have described hearing a certain musical note as a conical shape, tasting the name “Richard” as a warm, melting chocolate bar, seeing the sound of a woman’s voice as a thin sheet of metal, and hearing a cricket’s chirp as white. Synesthesia, by consensus definition, is when the presentation of a particular sensory stimulus elicits a perception in a second sensory modality without direct triggering of this second modality (1). This could be manifested as certain letters evoking colors, sounds having tastes, and so on. Because much of the research on synesthesia is going on presently and many theories are still being challenged, I am using this paper to explore the implications of research on synesthesia, not necessarily certain conclusions that can be drawn. My main goal is to explore what phenomena are occurring in the synesthetic brain that result in the synesthetic experience, and what this cause can tell us about the mechanisms of perception in general using synesthesia as a window into how the brain processes all types of information and stimuli. A secondary question raised to complement the “how” part of the developmental process of synesthesia is the “why”-there an underlying evolutionary benefit for not being synesthetic, and what does this say about how much of outer reality is useful for humans to be able to perceive? That will be left to the reader to consider after presentation of the brain-based evidence of the mechanisms of synesthesia, but is enlightening to keep in the back of the mind. However, I will aim to present as comprehensive a discussion as possible of the current evidence and theories concerning the roots and nature of synesthesia.

Synesthetic Reality
The sensory cross-activation in synesthetes can be brought on by many different inducers resulting in different types of secondary sensory experiences.

The most common type of synesthesia is “grapheme-color” synesthesia, in which specific letters are seen as certain colors. There have been many studies done to show that synesthesia is separate from simple learned association between stimuli, e.g., a child learning to associate blue with the letter D because of his refrigerator magnets or coloring books. For instance, when synesthetes were asked the colors they associate with letters, they had a 93 percent consistency in their responses when tested again a year later. However, nonsynesthetes had less than a 38 percent consistency when tested only a week later, even though they had been told to expect the retest (2).

Establishing Synesthetic Ability

Various methods have been used to determine who, exactly, is synesthetic, and to what degree. The most prominent model is the Stroop test, testing for the degree of interference when subjects are shown letters in colors incompatible with their synesthetic expectations. Just as normal subjects have difficulty reading a color word that is presented in a separate color than the meaning, synesthetes have slower reaction times when asked to name the color of a letter or digit if the color is incompatible with the color usually associated for them. Not only does the incompatibility cause longer reaction times, there is also measurable pupil dilation in the synesthetes as they examine the colored graphemes. In the way that those who take the Stroop test are unable to stop their processing of the meaning of the word, synesthetic subjects report their lack of ability to curb their incompatible color arising when they view the grapheme. Even more interestingly, the Stroop interference is most significant when the synesthete’s expected color for the grapheme is the complement of the color that actually appears in front of them. This finding suggests that synesthetic color arises from the same complementary color processes in the primary visual cortex that facilitate normal color vision.

Synthesthetes may mix number and letter cues with different colors as shown above.

Synthesthetes may mix number and letter cues with different colors as shown above.

Another method used to test synesthetic capacity was developed by Ramachandran and Hubbard. They designed a sort of “letter soup” image made up of scattered graphemes that contained a hidden embedded shape (3).

The image was flashed for one second in front of subjects, who were than asked what shape had been present. Synesthetes were significantly better at the task than non-synesthetes and had a much better ability to experience the “pop-out” effect of the embedded grapheme shape because of the colorful distinctions between the letters that appeared to them (4). Tests are also able to distinguish between higher versus lower synesthetes, which I will explore later in the paper.

In the Synesthetic Brain
Much of the recent research has used various brain imaging techniques to observe the brain activity of synesthetes. As mentioned before, imaging has shown that both distinct areas of the brain used for processing a specific sense are activated in a synesthete, e.g., for someone who hears words as colored, both their auditory cortex and the extrastriate visual cortex are activated upon listening to someone speaking. The brain activity seems to quantitatively relate to the level of synesthetic ability, as those synesthetes who scored highest on the “pop-out” visual tests showed the most neuronal activity in the color area when viewing sensory inducers.

Brain imaging data lends credence to the notion that the neuromechanisms facilitating synesthesia are similar to those that facilitate normal perception. The implications of this brain activity data have to do with the focal idea that the root of synesthesia lies in unusual communication in the brain, especially through connections that are not normally present or pathways that are not normally used. During a study on grapheme-color synesthesia, Ramachandran and Hubbard found that the V4 color area and the visual area that facilitated grapheme identification were both in the fusiform gyrus, and so they hypothesized that if neurons from the two areas began to communicate, the outcome would be a simultaneous experience of colors when seeing graphemes. On a similar hypothesis, if connections between normally separate cortical areas were created or activated, this could suggest a similar mechanism as in amputee patients with phantom limb sensations. In amputee patients, a reorganization of the cortex results in tactile sensations on a limb that does not exist anymore, so perhaps a similar local cross-activation model (Ramachandran and Hubbard) would account for synesthesia. In this way, the brain images provide both a launching point from which to start speculation about the roots of synesthesia as well as a reference point to come back to in order to test existing hypotheses.

The Causes and Development of Synesthesia

The prominent debate in attempts to figure out the cause of synesthesia can be reduced to two major hypotheses. The first is that synesthesia occurs because of cross-wiring between contiguous brain areas, either because of a deficiency of the proper synaptic pruning that would separate the areas or because of an excess of connections due to other reasons. The second hypothesis promotes a disinhibited feedback theory, the idea that there is excess activity between the levels of the sensory hierarchy or concurrent pathways because of a disinhibition of the feedback signals that would normally occur. These differences can thus come down to a structural cause, i.e., the extra connections of the first hypothesis, or a functional cause, i.e., the disinhibition of normally existing connections. Strong evidence for both theories makes the debate between the two that much more intriguing.

Cross-activation Due to Lack of Pruning
The cross-wiring hypothesis is based on the idea that adjacent brain areas do not get separated due to a mutation in the gene that would normally regulate the synaptic pruning process. In this way, the cross-activation is due to “extra wires,” to neural connections that were not pruned away. These crossed wires would then lead to a concurrent sense being experienced upon the occurrence of a separate inducer. There is plenty of evidence for the idea that young humans and other organisms have short-lived synaptic connections that are eventually pruned away with maturation. In an infant, the many connections make its cortex barely able to function, proven by low levels of blood flow and bad results at behavioral marker tasks (5). The pruning is a necessary process to allow normal functioning within the world, not necessarily a hampering of ability.

In the adult brain, each sensory area is focused on processing information from a different sensory modality. As stated earlier, studies have shown that synesthetes have activation in both of the cortical areas that are eliciting responses, e.g., for someone who hears words as colored, both the auditory cortex and the visual cortex are activated upon listening to someone speaking. The foundation of the cross-activation theory is that these sensory cortical areas are not initially as specialized as they become later in life. The synapses are pruned over maturation in an experience-dependent manner, and the over-wiring in infants that results in transient connections between adjacent brain areas suggests that these connections are in fact functional before they are pruned.

Genetic evidence could lend credibility to the pruning hypothesis. Synesthesia has been demonstrated to run in families and also to occur more in women than in men. This evidence would suggest that there is a chromosomal component to synesthesia, and that it is X-linked because of its prevalence in women. A genetic basis would neatly explain the pruning hypothesis, since a mutation of the gene that initiates the normal pruning process would have the direct result of extra synapses between cortical areas. The ratio in women to men has been found to be 6:1, although there is recent data challenging that. Regardless, the solid evidence for familial trends of synesthesia is a telling sign of its heritability factor (6,7).

Disinhibition of Feedback

The disinhibition theory posits a functional difference in synesthetes as opposed to a structural; that is, that a disinhibition of normal connections is the root of synesthesia. This hypothesis states that in synesthetes, there is an inhibition failure between adjacent cortical regions that are normally insulated from each other, and this inhibition failure causes concurrent sensory pathways to be activated along with the normal inducer ones.

Figure 1: Neural Network Model

Figure 1: Neural Network Model

In this case there are no abnormal horizontal neural connections, as would occur with a lack of pruning, but instead a lack of inhibition of top-down signaling via the feedback connections. As seen in Figure 1, feedforward and feedback connections are usually separate and differentiated based on where they originate and terminate. However, a failure of inhibition could lead to the activation of another pathway that normally remains independent.

In a nonsynesthetic adult, feedback from higher cortical areas onto lower sensory areas remains hierarchically organized through modules, as seen in Figure 1. The sensory stimulus at the bottom left stimulates the inducer pathway, which eventually leads to the resulting thought. An expected stimulus would, in the typical adult, fire neurons consistent with the end thought normally triggered by the inducer while inhibiting firing of other neurons that are not part of the inducer pathway. However, the disinhibition theory suggests that in synesthetes, some of the inhibitory feedback is disinhibited, which would allow concurrent pathways to activate along with the inducer pathway and thus result in a different brain area being activated by input from the “wrong” sense. As Spector and Maurer note, this theory rests on the assumption that adjacent areas of the cortex are always connected with each other, but that the connections are normally functionally inhibited.

Are We All Synesthetes as Infants? Do We Still Have this Potential?

Neonatal Synesthetic Responses
The theory exists that we all might have synesthesia-like abilities as infants, but lose them as we age. In essence, the neonatal synesthesia model and the cross-modal transfer hypothesis theories propose that newborn infants are unable to differentiate input from their different senses, and have much cross-talk between diffuse sensory regions. A great deal of the recent evidence combines data on developmental plasticity and the neuronal activity of both human synesthetes and neonates of other species.

Much evidence exists to suggest that young organisms of all species have a surplus of synaptic connections between cortical areas that will be pruned as they mature. Data also implies that the cortex of a newborn infant is barely functioning in comparison to that of an adult, as there has been found to be low levels of blood flow, anatomical immaturities, and poor performance at developmental tasks. On the other hand, the infant’s limbic system appears to develop and function well much earlier in comparison to the cortex, likely even being functional at birth. There are thus a few possibilities for synesthetic ability in infants; some of the transient connections between cortical areas that will eventually become pruned might still be operative, the limbic system is functioning with little input from the cortex, and/or perhaps due to the immature cortex, a baby does not differentiate stimuli from distinct modalities (8).

There is intriguing evidence for the lack of sense separation in infants. Molina and Jouen did a study in 2001 linking infants’ perception of texture with their response to a visual cue presented as well. They first found that infants preferred to squeeze smooth objects over granular-textured ones, and then presented infants with visual stimuli while they were squeezing the objects. The visual stimuli would either match the texture of the object (e.g., textured object and textured visual stimuli), or would differ. They found that when a matching visual stimulus was presented the infants’ rate of squeezing remained the same, but that when they presented a discordant visual image the infants squeezing frequency suddenly changed. This parallels the findings described earlier about pupil dilation in grapheme-color synesthetes, how the subjects showed increased pupil dilation when presented with an image of a letter in the opposite color than they would normally perceive. Because of this, Maurer and Mondlach assert that the results of the infants signify a merging of the senses in the newborns and preferential mappings from one sensory modality to the other (9).

Synesthetic Potential in All of Us?

Figures used to test the Kiki-Bouba Effect

Figures used to test the Kiki-Bouba Effect

The data on the possibility of synesthetic capability in infants leads to the next question-is there then still the foundational capacity for synesthesia in all of us, only manifested explicitly in synesthetes?
Can you tell who is Kiki and who is Bouba? When this image is shown to nonsynesthetes, nearly 100 percent of them identify the one on the left as Kiki and the one on the right as Bouba. The question then becomes whether this is a demonstration of deeply ingrained synesthetic tendencies in all adults to be able to connect visual stimuli with auditory correlates, or whether there is simply cultural or some other developmental tendency expressed in this example. Ward concludes that this demonstrates an innate ability wired into all our brains to recognize the associations that synesthetes perceive. In regard to the link between synesthesia and increased creativity, Ward further expands these findings to speculate, “what it can tell us about how we appreciate art, and why certain art forms are appealing. We’re starting to use principles of neuroscience to understand what artists have been doing for centuries” (10,11). Cytowic, a leading synesthesia researcher, at one point asserted, “Synesthesia is actually a normal brain function in every one of us, but its workings reach conscious awareness in only a handful” (12). Does synesthesia indeed reflect a basic property of the brain? This would be supported more by the disinhibition theory, since in that case we would all have the necessary structural wiring but it would be more about activating pathways that are normally inhibited.

Robertson and Sagiv succinctly illuminate the wider implications of synesthesia: It may even question fundamental assumptions about the nature of biological systems…to philosophical questions of functionalism…it is a scientific puzzle, and it raises fundamental issues about how it may relate to “normal” perception and how brains must work such that they can generate such phenomena. (13)

Synesthesia is a fascinating puzzle that highlights many angles of current brain research. From the mechanisms of perception to the developmental processes of infants, looking at how synesthetes interact with their daily reality tells us more about those of us without the ability as well. Looking at the link between synesthesia and creative ability, especially through its prevalence in the literary and artistic professional world, one wonders at the enhanced possibilities gained from seeing through the synesthetic lens and the perceptive results if we were able to intentionally cultivate synesthesia.

By highlighting the functional and developmental processes of the brain, the process of synesthesia sheds new light on the subtleties of perception and neuroplasticity. The intersection of these two areas is a ripe place for further research, and encourages exploration of the ability of the brain to take over senses, regenerate damaged senses, and perhaps even promote new ones. Ultimately, synesthesia also brings to the forefront the eternal debate about how much of what we perceive as reality is a unique perception based on our individual abilities, or how much is a stable entity that retains its essential identity regardless of the way in which it is perceived. In this sense, synesthesia can be viewed as a valuable lens through which to further explain the nature of our brains, our surrounding environment, and how our perception shapes ourselves.


1. S. Baron-Cohen, J. Harrison, in Neurodevelopmental Disorders, H. Tager-Flusberg, Ed. (MIT Press, Cambridge, MA, 1999).
2. F. Spector, D. Maurer, Synesthesia: A New Approach to Understanding the Development of Perception. Dev. Psychol. 45.1, 175-89 (2009).
3. . V.S. Ramachandran, E.M. Hubbard. Journal of Consciousness Studies, 8(12), 3-34 (Dec 2001).
4. M. Hochel, E.G. Milán, Synaesthesia: The existing state of affairs. Cognitive Neuropsychology. 25(1), 93-117 (Feb. 2008).
5. D. Maurer, C. J. Mondloch, in Attention on synesthesia: Cognition, development and neuroscience, eds. L. Robertson, N. Sagiv, (Oxford: Oxford University Press, 1998), 193-213.
6. G. Bargary, K.J. Mitchell, Trends Neurosci. 31(11), (Nov. 2008):
7. E.M. Hubbard, V.S. Ramachandran, Journal of Consciousness Studies. 10(3), 77-84 (Mar. 2003).
8. R.C. Kadosh, A. Henik, V. Walsh, Developmental Science, 12(3), 484-91 (May 2009).
9. S. Baron-Cohen, PSYCHE, 2(27), (Jun. 1996).
10. V. Hughes, The most beautiful painting you’ve ever heard. Seed Magazine. December 13, 2006.
11. J. Ward, J.B.Mattingley. Cortex, 42(2), 129-36 (Feb. 2006).
12. R. Walsh, Journal of Consciousness Studies. 12.4-5, 12 (2005).
13. L. Robertson, N. Sagiv, Eds., Synesthesia: Perspectives from Cognitive Neuroscience. (Oxford University Press, New York, 2005), p. vii.